ncBacktester 0.1.1

A simple backtesting engine for trading strategies

ncBacktester

ncBacktester is a small, opinionated Python package for backtesting simple long-only strategies driven by a Hold_Signal column on OHLCV data. It is designed for clarity and education rather than exhaustive feature coverage.

Quick summary: feed time-series OHLCV data with a Hold_Signal (1 = hold, 0 = not hold), instantiate Backtest, call .run() and examine metrics and plots.

Install (development)

pip install ncBacktester

Minimal Quick Start

import pandas as pd
from ncBacktester.backtest import Backtest

# data must include: Open, High, Low, Close, Volume, Hold_Signal
data = pd.read_csv('your_data.csv', parse_dates=['Date'], index_col='Date')

bt = Backtest(data=data, initial_capital=10000.0, stop_loss_pct=0.05, commission=0.001)
results = bt.run()

print(results['metrics'])          # key performance metrics
results['equity_curve'].plot()     # equity curve
bt.plot(save_path='backtest.png')  # static plots

Required data format

• Columns: Open, High, Low, Close, Volume, Hold_Signal.
• Hold_Signal should be 0 or 1; changes 0→1 trigger buys, 1→0 trigger sells.

Main components

• Backtest (ncBacktester/ncBacktester/backtest.py): orchestrates the workflow — validate data, initialize components, execute strategy, compute metrics, and prepare plots.
• StrategyExecutor (ncBacktester/ncBacktester/strategy_executor.py): implements trade sizing and execution logic (uses Close prices and accounts for commission).
• MetricsCalculator (ncBacktester/ncBacktester/metrics.py): computes performance statistics (Sharpe, Sortino, CAGR, drawdowns, etc.) from trades and the equity curve.
• StopLossManager (ncBacktester/ncBacktester/stop_loss.py): supports fixed and trailing stop loss behaviour during the backtest.
• Plotter (ncBacktester/ncBacktester/plotter.py): generates static plots showing price, trades, equity curve and drawdowns.

Behavior & assumptions

• Buys and sells are executed at the bar Close where the Hold_Signal changes.
• On a buy the framework purchases as many shares as possible with available capital; on sell it liquidates the full position.
• Stop loss (fixed or trailing) is optional and configured via Backtest(stop_loss_pct=..., trailing_stop_pct=...).

Advanced usage

• If you want to reuse StrategyExecutor or StopLossManager directly, import them from their modules and instantiate with the same parameters used by Backtest.

Contributing & license

• This project is intended for learning; contributions and issues are welcome.
• Licensed under the MIT License.